miketranz 969 posts msg #122507 - Ignore miketranz |
1/11/2015 9:57:24 AM
If the SPY makes a 10 day low and the amount of stocks that come up on the filter is 1200+,what's the percentage of times the market will go up the following 1-7 days,using a 1 year back test.Also,what would be the optimum short term hold in days.Thanks,Miketranz........
|
Eman93 4,750 posts msg #122522 - Ignore Eman93 modified |
1/15/2015 12:46:13 AM
Mike this is what I have been wanting for years here, this could be something revolutionary SF could create.
All I want is a simple plot of the number of stocks MY FILTER returned over 3 years, hey that would be a nice start.
The back test app is doing the calculations and running the filter over the time period any way, just add a few lines of code and get to spit out the date and number of hits., it doesn't seem like a big deal. hell I will take a simple report: Date and number of hits.
Just think you can make you own custom market indicator, just scanning for stocks is only part of trading.. think McClellan summation. now that would be very analytical.
I have a few filters and if they get an unusual number of hits I know its going to be a nice general market move.
|
miketranz 969 posts msg #122524 - Ignore miketranz |
1/15/2015 8:56:38 AM
Eman,yes,we're on the same page.I'm just trying to run a set of market conditions,to see whether the stats are profitable,and to what percent profitable the following days.I've seen this pattern work consistently.I just want to know what the real numbers are.Are there any sites on the web where you can input this kind of data? Maybe Kevin can shed some kind of light on this.Thanks,Miketranz...
|
Kevin_in_GA 4,599 posts msg #122531 - Ignore Kevin_in_GA |
1/15/2015 1:47:28 PM
A lot of it depends on what type of filter you are using, and the market to which it is applied. For example, if one looks at a 10 day low in the SPY I would probably use only the S&P 500 stocks, or perhaps extend it to the Russell 2000, and not look at penny stocks.
Stratasearch has a function that could work here - numtotal(). It counts the number of stocks within a defined set like the S&P500 that have met a specific criterion, and can be used in filters. Example - you might want the percentage of S&P500 stocks that closed below their lower Bollinger band(20,2) to be part of a filter's entry criteria, so it would look like
numtotal("S&P-500", close < lbb(close,20,2))/numtotal("S&P-500", close > 0.01)
|
miketranz 969 posts msg #122535 - Ignore miketranz |
1/16/2015 8:52:59 AM
Can a 10 day low indicator be incorporated into a SPY chart w/number of stocks that hit the filter.I'm basically trying to find a short term low reversal swing trade.What works well is the Connors all in one,just trying to tweek it a bit if possible.Thanks again,Mike...
|
four 5,087 posts msg #122536 - Ignore four modified |
1/16/2015 1:43:48 PM
http://www.crestmontresearch.com/docs/Stock-Yo-Yo.pdf
PERCENTAGE POSITIVE AND NEGATIVE DAYS ACROSS VARIOUS PERIODS: S&P 500 INDEX
--- ---
http://prototypes.stockmapper.com/NYXtrac.html
Found this... ?? Click on the "box" to change to the item listed. Boxes are at the top of the page and act like buttons.
|
miketranz 969 posts msg #122541 - Ignore miketranz |
1/17/2015 8:52:07 AM
Thanks for all the help,appreciate it.Miketranz...
|
Eman93 4,750 posts msg #122552 - Ignore Eman93 modified |
1/18/2015 10:10:38 PM
Kevin_in_GA
3,411 posts
msg #122531
- Ignore Kevin_in_GA 1/15/2015 1:47:28 PM
A lot of it depends on what type of filter you are using, and the market to which it is applied. For example, if one looks at a 10 day low in the SPY I would probably use only the S&P 500 stocks, or perhaps extend it to the Russell 2000, and not look at penny stocks.
Stratasearch has a function that could work here - numtotal(). It counts the number of stocks within a defined set like the S&P500 that have met a specific criterion, and can be used in filters. Example - you might want the percentage of S&P500 stocks that closed below their lower Bollinger band(20,2) to be part of a filter's entry criteria, so it would look like
numtotal("S&P-500", close < lbb(close,20,2))/numtotal("S&P-500", close > 0.01)
================================================================================
If thats the case we can all be rich!
I will check it out
|